Engineering Acoustics/Print version
This is the print version of Engineering_Acoustics You won't see this message or any elements not part of the book's content when you print or preview this page. 
Note: current version of this book can be found at http://en.wikibooks.org/wiki/Engineering_Acoustics
Remember to click "refresh" to view this version.
Part 1: Lumped Acoustical Systems
Simple Oscillation
The Position Equation
This section shows how to form the equation describing the position of a mass on a spring.
For a simple oscillator consisting of a mass m attached to one end of a spring with a spring constant s, the restoring force, f, can be expressed by the equation
where x is the displacement of the mass from its rest position. Substituting the expression for f into the linear momentum equation,
where a is the acceleration of the mass, we can get
or,
Note that the frequency of oscillation is given by
To solve the equation, we can assume
The force equation then becomes
Giving the equation
Solving for
This gives the equation of x to be
Note that
and that C_{1} and C_{2} are constants given by the initial conditions of the system
If the position of the mass at t = 0 is denoted as x_{0}, then
and if the velocity of the mass at t = 0 is denoted as u_{0}, then
Solving the two boundary condition equations gives
The position is then given by
This equation can also be found by assuming that x is of the form
And by applying the same initial conditions,
This gives rise to the same position equation
Alternate Position Equation Forms
If A_{1} and A_{2} are of the form
Then the position equation can be written
By applying the initial conditions (x(0)=x_{0}, u(0)=u_{0}) it is found that
If these two equations are squared and summed, then it is found that
And if the difference of the same two equations is found, the result is that
The position equation can also be written as the Real part of the imaginary position equation
Due to euler's rule (e^{jφ} = cosφ + jsinφ), x(t) is of the form
Forced Oscillations(Simple SpringMass System)
Recap of Section 1.3
In the previous section, we discussed how adding a damping component (e. g. a dashpot) to an unforced, simple springmass system would affect the response of the system. In particular, we learned that adding the dashpot to the system changed the natural frequency of the system from to a new damped natural frequency , and how this change made the response of the system change from a constant sinusoidal response to an exponentiallydecaying sinusoid in which the system either had an underdamped, overdamped, or criticallydamped response.
In this section, we will digress a bit by going back to the simple (undamped) oscillator system of the previous section, but this time, a constant force will be applied to this system, and we will investigate this system’s performance at low and high frequencies as well as at resonance. In particular, this section will start by introducing the characteristics of the spring and mass elements of a springmass system, introduce electrical analogs for both the spring and mass elements, learn how these elements combine to form the mechanical impedance system, and reveal how the impedance can describe the mechanical system’s overall response characteristics. Next, power dissipation of the forced, simple springmass system will be discussed in order to corroborate our use of electrical circuit analogs for the forced, simple springmass system. Finally, the characteristic responses of this system will be discussed, and a parameter called the amplification ratio (AR) will be introduced that will help in plotting the resonance of the forced, simple springmass system.
Forced Spring Element
Taking note of Figs. 1, we see that the equation of motion for a spring that has some constant, external force being exerted on it is...
where is the mechanical stiffness of the spring.
Note that in Fig. 1(c), force flows constantly (i.e. without decreasing) throughout a spring, but the velocity of the spring decrease from to as the force flows through the spring. This concept is important to know because it will be used in subsequent sections.
In practice, the stiffness of the spring , also called the spring constant, is usually expressed as , or the mechanical compliance of the spring. Therefore, the spring is very stiff if is large is small. Similarly, the spring is very loose or “bouncy” if is small is large. Noting that force and velocity are analogous to voltage and current, respectively, in electrical systems, it turns out that the characteristics of a spring are analogous to the characteristics of a capacitor in relation to, and, so we can model the “reactiveness” of a spring similar to the reactance of a capacitor if we let as shown in Fig. 2 below.
Forced Mass Element
Taking note of Fig. 3, the equation for a mass that has constant, external force being exerted on it is...
If the mass can vary its value and is oscillating in a mechanical system at max amplitude such that the input the system receives is constant at frequency , as increases, the harder it will be for the system to move the mass at at until, eventually, the mass doesn’t oscillate at all . Another equivalently way to look at it is to let vary and hold constant. Similarly, as increases, the harder it will be to get to oscillate at and keep the same amplitude until, eventually, the mass doesn’t oscillate at all. Therefore, as increases, the “reactiveness” of mass decreases (i.e. starts to move less and less). Recalling the analogous relationship of force/voltage and velocity/current, it turns out that the characteristics of a mass are analogous to an inductor. Therefore, we can model the “reactiveness” of a mass similar to the reactance of an inductor if we let as shown in Fig. 4.
Mechanical Impedance of SpringMass System
As mentioned twice before, force is analogous to voltage and velocity is analogous to current. Because of these relationships, this implies that the mechanical impedance for the forced, simple springmass system can be expressed as follows:
In general, an undamped, springmass system can either be “springlike” or “masslike”. “Springlike” systems can be characterized as being “bouncy” and they tend to grossly overshoot their target operating level(s) when an input is introduced to the system. These type of systems relatively take a long time to reach steadystate status. Conversely, “masslike” can be characterized as being “lethargic” and they tend to not reach their desired operating level(s) for a given input to the system...even at steadystate! In terms of complex force and velocity, we say that “ force LEADS velocity” in masslike systems and “velocity LEADS force” in springlike systems (or equivalently “ force LAGS velocity” in masslike systems and “velocity LAGS force” in springlike systems). Figs. 5 shows this relationship graphically.
Power Transfer of a Simple SpringMass System
From electrical circuit theory, the average complex power dissipated in a system is expressed as ...
where and represent the (timeinvariant) complex voltage and complex conjugate current, respectively. Analogously, we can express the net power dissipation of the mechanical system in general along with the power dissipation of a springlike system or masslike system as...
In equations 1.4.7, we see that the product of complex force and velocity are purely imaginary. Since reactive elements, or commonly called, lossless elements, cannot dissipate energy, this implies that the net power dissipation of the system is zero. This means that in our simple springmass system, power can only be (fully) transferred back and forth between the spring and the mass. But this is precisely what a simple springmass system does. Therefore, by evaluating the power dissipation, this corroborates the notion of using electrical circuit elements to model mechanical elements in our springmass system.
Responses For Forced, Simple SpringMass System
Fig. 6 below illustrates a simple springmass system with a force exerted on the mass.
This system has response characteristics similar to that of the undamped oscillator system, with the only difference being that at steadystate, the system oscillates at the constant force magnitude and frequency versus exponentially decaying to zero in the unforced case. Recalling equations 1.4.2b and 1.4.4b, letting be the natural (resonant) frequency of the springmass system, and letting be frequency of the input received by the system, the characteristic responses of the forced springmass systems are presented graphically in Figs. 7 below.
Amplification Ratio
The amplification ratio is a useful parameter that allows us to plot the frequency of the springmass system with the purports of revealing the resonant freq of the system solely based on the force experienced by each, the spring and mass elements of the system. In particular, AR is the magnitude of the ratio of the complex force experienced by the spring and the complex force experienced by the mass, i.e.
If we let , be the frequency ratio, it turns out that AR can also be expressed as...
AR will be at its maximum when . This happens precisely when . An example of an AR plot is shown below in Fig 8.
Mechanical Resistance
Mechanical Resistance
For most systems, a simple oscillator is not a very accurate model. While a simple oscillator involves a continuous transfer of energy between kinetic and potential form, with the sum of the two remaining constant, real systems involve a loss, or dissipation, of some of this energy, which is never recovered into kinetic nor potential energy. The mechanisms that cause this dissipation are varied and depend on many factors. Some of these mechanisms include drag on bodies moving through the air, thermal losses, and friction, but there are many others. Often, these mechanisms are either difficult or impossible to model, and most are nonlinear. However, a simple, linear model that attempts to account for all of these losses in a system has been developed.
Dashpots
The most common way of representing mechanical resistance in a damped system is through the use of a dashpot. A dashpot acts like a shock absorber in a car. It produces resistance to the system's motion that is proportional to the system's velocity. The faster the motion of the system, the more mechanical resistance is produced.
As seen in the graph above, a linear relationship is assumed between the force of the dashpot and the velocity at which it is moving. The constant that relates these two quantities is , the mechanical resistance of the dashpot. This relationship, known as the viscous damping law, can be written as:
Also note that the force produced by the dashpot is always in phase with the velocity.
The power dissipated by the dashpot can be derived by looking at the work done as the dashpot resists the motion of the system:
Modeling the Damped Oscillator
In order to incorporate the mechanical resistance (or damping) into the forced oscillator model, a dashpot is placed next to the spring. It is connected to the mass () on one end and attached to the ground on the other end. A new equation describing the forces must be developed:
It's phasor form is given by the following:
Mechanical Impedance for Damped Oscillator
Previously, the impedance for a simple oscillator was defined as . Using the above equations, the impedance of a damped oscillator can be calculated:
For very low frequencies, the spring term dominates because of the relationship. Thus, the phase of the impedance approaches for very low frequencies. This phase causes the velocity to "lag" the force for low frequencies. As the frequency increases, the phase difference increases toward zero. At resonance, the imaginary part of the impedance vanishes, and the phase is zero. The impedance is purely resistive at this point. For very high frequencies, the mass term dominates. Thus, the phase of the impedance approaches and the velocity "leads" the force for high frequencies.
Based on the previous equations for dissipated power, we can see that the real part of the impedance is indeed . The real part of the impedance can also be defined as the cosine of the phase times its magnitude. Thus, the following equations for the power can be obtained.
Characterizing Damped Mechanical Systems
Characterizing Damped Mechanical Systems
Characterizing the response of Damped Mechanical Oscillating system can be easily quantified using two parameters. The system parameters are the resonance frequency ( and the damping of the system . In practice, finding these parameters would allow for quantification of unknown systems and allow you to derive other parameters within the system.
Using the mechanical impedance in the following equation, notice that the imaginary part will equal zero at resonance.
()
Resonance case:()
Calculating the Mechanical Resistance
The decay time of the system is related to 1 / B where B is the Temporal Absorption. B is related to the mechancial resistance and to the mass of the system by the following equation.
The mechanical resistance can be derived from the equation by knowing the mass and the temporal absorption.
Critical Damping
The system is said to be critically damped when:
A critically damped system is one in which an entire cycle is never completed. The absorption coefficient in this type of system equals the natural frequency. The system will begin to oscillate, however the amplitude will decay exponentially to zero within the first oscillation.
Damping Ratio
The damping ratio is a comparison of the mechanical resistance of a system to the resistance value required for critical damping. Rc is the value of Rm for which the absorption coefficient equals the natural frequency (critical damping). A damping ratio equal to 1 therefore is critically damped, because the mechanical resistance value Rm is equal to the value required for critical damping Rc. A damping ratio greater than 1 will be overdamped, and a ratio less than 1 will be underdamped.
?????
Quality Factor
The Quality Factor (Q) is way to quickly characterize the shape of the peak in the response. It gives a quantitative representation of power dissipation in an oscillation.
Wu and Wl are called the half power points. When looking at the response of a system, the two places on either side of the peak where the point equals half the power of the peak power defines Wu and Wl. The distance in between the two is called the halfpower bandwidth. So, the resonant frequency divided by the halfpower bandwidth gives you the quality factor. Mathematically, it takes Q/pi oscillations for the vibration to decay to a factor of 1/e of its original amplitude.
ElectroMechanical Analogies
Why Circuit Analogs?
Acoustic devices are often combinations of mechanical and electrical elements. A common example of this would be a loudspeaker connected to a power source. It is useful in engineering applications to model the entire system with one method. This is the reason for using a circuit analogy in a vibrating mechanical system. The same analytic method can be applied to ElectroAcoustic Analogies.
How ElectroMechanical Analogies Work
An electrical circuit is described in terms of its potential (voltage) and flux (current). To construct a circuit analog of a mechanical system we define flux and potential for the system. This leads to two separate analog systems. The Impedance Analog denotes the force acting on an element as the potential and the velocity of the element as the flux. The Mobility Analog equates flux with the force and velocity with potential.
Mechanical  Electrical Equivalent  

Impedance Analog  
Potential:  Force  Voltage 
Flux:  Velocity  Current 
Mobility Analog  
Potential:  Velocity  Voltage 
Flux:  Force  Current 
For many, the mobility analog is considered easier for a mechanical system. It is more intuitive for force to flow as a current and for objects oscillating the same frequency to be wired in parallel. However, either method will yield equivalent results and can also be translated using the dual (dot) method.
The Basic Elements of an Oscillating Mechanical System
The Mechanical Spring:
The ideal spring is considered to be operating within its elastic limit, so the behavior can be modeled with Hooke's Law. It is also assumed to be massless and have no damping effects.
The Mechanical Mass
In a vibrating system, a mass element opposes acceleration. From Newton's Second Law:
The Mechanical Resistance
The dashpot is an ideal viscous damper which opposes velocity.
Ideal Generators
The two ideal generators which can drive any system are an ideal velocity and ideal force generator. The ideal velocity generator can be denoted by a drawing of a crank or simply by declaring , and the ideal force generator can be drawn with an arrow or by declaring
Simple Damped Mechanical Oscillators
In the following sections we will consider this simple mechanical system as a mobility and impedance analog. It can be driven either by an ideal force or an ideal velocity generator, and we will consider simple harmonic motion. The m in the subscript denotes a mechanical system, which is currently redundant, but can be useful when combining mechanical and acoustic systems.
The Impedance Analog
The Mechanical Spring
In a spring, force is related to the displacement from equilibrium. By Hooke's Law,
The equivalent behaviour in a circuit is a capacitor:
The Mechanical Mass
The force on a mass is related to the acceleration (change in velocity). The behaviour, by Newton's Second Law, is:
The equivalent behaviour in a circuit is an inductor:
The Mechanical Resistance
For a viscous damper, the force is directly related to the velocity
The equivalent is a simple resistor of value
Example:
Thus the simple mechanical oscillator in the previous section becomes a series RCL Circuit:
The current through all three elements is equal (they are at the same velocity) and that the sum of the potential drops across each element will equal the potential at the generator (the driving force). The ideal voltage generator depicted here would be equivalent to an ideal force generator.
IMPORTANT NOTE: The velocity measured for the spring and dashpot is the relative velocity ( velocity of one end minus the velocity of the other end). The velocity of the mass, however, is the absolute velocity.
Impedances:
Element  Impedance  

Spring  Capacitor  
Mass  Inductor  
Dashpot  Resistor 
The Mobility Analog
Like the Impedance Analog above, the equivalent elements can be found by comparing their fundamental equations with the equations of circuit elements. However, since circuit equations usually define voltage in terms of current, in this case the analogy would be an expression of velocity in terms of force, which is the opposite of convention. However, this can be solved with simple algebraic manipulation.
The Mechanical Spring
The equivalent behavior for this circuit is the behavior of an inductor.
The Mechanical Mass
Similar to the spring element, if we take the general equation for a capacitor and differentiate,
The Mechanical Resistance
Since the relation between force and velocity is proportionate, the only difference is that the mechanical resistance becomes inverted:
Example:
The simple mechanical oscillator drawn above would become a parallel RLC Circuit. The potential across each element is the same because they are each operating at the same velocity. This is often the more intuitive of the two analogy methods to use, because you can visualize force "flowing" like a flux through your system. The ideal voltage generator in this drawing would correspond to an ideal velocity generator.
IMPORTANT NOTE: Since the measure of the velocity of a mass is absolute, a capacitor in this analogy must always have one terminal grounded. A capacitor with both terminals at a potential other than ground may be realized physically as an inverter, which completes all elements of this analogy.
Impedances:
Element  Impedance  

Spring  Inductor  
Mass  Capacitor  
Dashpot  Resistor 
Methods for checking ElectroMechanical Analogies
After drawing the electromechanical analogy of a mechanical system, it is always safe to check the circuit. There are two methods to accomplish this:
Review of Circuit Solving Methods
Kirchkoff's Voltage law
"The sum of the potential drops around a loop must equal zero."
Kirchkoff's Current Law
"The Sum of the currents at a node (junction of more than two elements) must be zero"
Hints for solving circuits:
Remember that certain elements can be combined to simplify the circuit (the combination of like elements in series and parallel)
If solving a ciruit that involves steadystate sources, use impedances. Any circuit can eventually be combined into a single impedance using the following identities:
Impedances in series:
Impedances in parallel:
Dot Method: (Valid only for planar network)
This method helps obtain the dual analog (one analog is the dual of the other). The steps for the dot product are as follows: 1) Place one dot within each loop and one outside all the loops. 2) Connect the dots. Make sure that only there is only one line through each element and that no lines cross more than one element. 3) Draw in each line that crosses an element its dual element, including the source. 4) The circuit obtained should have an equivalent behavior as the dual analog of the original electromechanical circuit.
Example:
The parallel RLC Circuit above is equivalent to a series RLC driven by an ideal current source
LowFrequency Limits:
This method looks at the behavior of the system for very large or very small values of the parameters and compares them with the expected behavior of the mechanical system. For instance, you can compare the mobility circuit behavior of a nearinfinite inductance with the mechanical system behavior of a nearinfinite stiffness spring.
Very High Value  Very Low Value  

Capacitor  Short Circuit  Open Circuit 
Inductor  Open Circuit  Closed Circuit 
Resistor  Open Circuit  Short Circuit 
Additional Resources for solving linear circuits:
Thomas & Rosa, "The Analysis and Design of Linear Circuits", Wiley, 2001
Hayt, Kemmerly & Durbin, "Engineering Circuit Analysis", 6th ed., McGraw Hill, 2002
Examples of ElectroMechanical Analogies
Example System
Note: The crank indicates an ideal velocity generator, with an amplitude of rotating at rad/s.
Impedance Analog Solution
Mobility Analog Solution
Primary variables of interest
Basic Assumptions
Consider a piston moving in a tube. The piston starts moving at time t=0 with a velocity u=. The piston fits inside the tube smoothly without any friction or gap. The motion of the piston creates a planar sound wave or acoustic disturbance traveling down the tube at a constant speed c>>. In a case where the tube is very small, one can neglect the time it takes for acoustic disturbance to travel from the piston to the end of the tube. Hence, one can assume that the acoustic disturbance is uniform throughout the tube domain.
Assumptions
1. Although sound can exist in solids or fluid, we will first consider the medium to be a fluid at rest. The ambient, undisturbed state of the fluid will be designated using subscript zero. Recall that a fluid is a substance that deforms continuously under the application of any shear (tangential) stress.
2. Disturbance is a compressional one (as opposed to transverse).
3. Fluid is a continuum: infinitely divisible substance. Each fluid property assumed to have definite value at each point.
4. The disturbance created by the motion of the piston travels at a constant speed. It is a function of the properties of the ambient fluid. Since the properties are assumed to be uniform (the same at every location in the tube) then the speed of the disturbance has to be constant. The speed of the disturbance is the speed of sound, denoted by letter with subscript zero to denote ambient property.
5. The piston is perfectly flat, and there is no leakage flow between the piston and the tube inner wall. Both the piston and the tube walls are perfectly rigid. Tube is infinitely long, and has a constant area of cross section, A.
6. The disturbance is uniform. All deviations in fluid properties are the same across the tube for any location x. Therefore the instantaneous fluid properties are only a function of the Cartesian coordinate x (see sketch). Deviations from the ambient will be denoted by primed variables.
Variables of interest
Pressure (force / unit area)
Pressure is defined as the normal force per unit area acting on any control surface within the fluid.
For the present case,inside a tube filled with a working fluid, pressure is the ratio of the surface force acting onto the fluid in the control region and the tube area. The pressure is decomposed into two components  a constant equilibrium component, , superimposed with a varying disturbance . The deviation is also called the acoustic pressure. Note that can be positive or negative. Unit: . Acoustical pressure can be measured using a microphone.
Density
Density is mass of fluid per unit volume. The density, ρ, is also decomposed into the sum of ambient value (usually around ρ0= 1.15 kg/m3) and a disturbance ρ’(x). The disturbance can be positive or negative, as for the pressure. Unit:
Acoustic volume velocity
Rate of change of fluid particles position as a function of time. Its the well known fluid mechanics term, flow rate.
In most cases, the velocity is assumed constant over the entire cross section (plug flow), which gives acoustic volume velocity as a product of fluid velocity and cross section S.
Electroacoustic analogies
Electroacoustical Analogies
Acoustical Mass
Consider a rigid tubepiston system as following figure.
Piston is moving back and forth sinusoidally with frequency of f. Assuming (where c is sound velocity ), volume of fluid in tube is,
Then mass (mechanical mass) of fluid in tube is given as,
For sinusoidal motion of piston, fluid move as rigid body at same velocity as piston. Namely, every point in tube moves with the same velocity.
Applying the Newton's second law to the following free body diagram,
Where, plug flow assumption is used.
"Plug flow" assumption:
Frequently in acoustics, the velocity distribution along the normal surface of
fluid flow is assumed uniform. Under this assumption, the acoustic volume velocity U is
simply product of velocity and entire surface.
Acoustical Impedance
Recalling mechanical impedance,
acoustical impedance (often termed an acoustic ohm) is defined as,
where, acoustical mass is defined.
Acoustical Mobility
Acoustical mobility is defined as,
Impedance Analog vs. Mobility Analog
Acoustical Resistance
Acoustical resistance models loss due to viscous effects (friction) and flow resistance (represented by a screen).
Acoustical Generators
The acoustical generator components are pressure, P and volume velocity, U, which are analogus to force, F and velocity, u of electromechanical analogy respectively. Namely, for impedance analog, pressure is analogus to voltage and volume velocity is analogus to current, and vice versa for mobility analog. These are arranged in the following table.
Impedance and Mobility analogs for acoustical generators of constant pressure and constant volume velocity are as follows:
Acoustical Compliance
Consider a piston in an enclosure.
When the piston moves, it displaces the fluid inside the enclosure. Acoustic compliance is the measurement of how "easy" it is to displace the fluid.
Here the volume of the enclosure should be assumed to be small enough that the fluid pressure remains uniform.
Assume no heat exchange 1.adiabatic 2.gas compressed uniformly , p prime in cavity everywhere the same.
from thermo equitation File:Equ1.jpg it is easy to get the relation between disturbing pressure and displacement of the piston File:Equ3.gif where U is volume rate, P is pressure according to the definition of the impendance and mobility, we can getFile:Equ4.gif
Mobility Analog VS Impedance Analog
Examples of ElectroAcoustical Analogies
Example 1: Helmholtz Resonator
Assumptions  (1) Completely sealed cavity with no leaks. (2) Cavity acts like a rigid body inducing no vibrations.
Solution:
Example 2: Combination of SideBranch Cavities
Solution:
Transducers  Loudspeaker
Acoustic transducer
The purpose of an acoustic transducer is to convert electrical energy into acoustic energy. Many variations of acoustic transducers exist, such as electrostatic, balanced armature and movingcoil loudspeakers. This article focuses on movingcoil loudspeakers since they are the most commonly used type of acoustic transducer. First, the physical construction and principle of a typical moving coil transducer are discussed briefly. Second, electromechanoacoustical modeling of each element composing the loudspeaker is presented in a tutorial way to reinforce and supplement the theory on electromechanical analogies and electroacoustic analogies previously seen in other sections. Third, the equivalent circuit is analyzed to introduce the theory behind ThieleSmall parameters, which are very useful when designing loudspeaker enclosures. A method to experimentally determine ThieleSmall parameters is also included.
Movingcoil loudspeaker construction and principle
The classic movingcoil loudspeaker driver can be divided into three key components:
1) The magnet motor drive system, comprising the permanent magnet, the center pole and the voice coil acting together to produce a mechanical force on the diaphragm from an electrical current.
2) The loudspeaker cone system, comprising the diaphragm and dust cap, permitting mechanical force to be translated into acoustic pressure;
3) The loudspeaker suspension, comprising the spider and surround, preventing the diaphragm from breaking due to over excursion, allowing only translational movement and tending to bring the diaphragm back to its rest position.
The following illustration shows a cutaway view of a typical moving coilpermanent magnet loudspeaker. A coil is mechanically coupled to a diaphragm, also called cone, and rests in a fixed magnetic field produced by a magnet. When an electrical current flows through the coil, a corresponding magnetic field is emitted, interacting with the fixed field of the magnet and thus applying a force to the coil, pushing it away or towards the magnet. Since the cone is mechanically coupled to the coil, it will push or pull the air it is facing, causing pressure changes and emitting a sound wave.
An equivalent circuit can be obtained to model the loudspeaker as a lumped system. This circuit can be used to drive the design of a complete loudspeaker system, including an enclosure and sometimes even an amplifier that is matched to the properties of the driver. The following section shows how such an equivalent circuit can be obtained.
Electromechanoacoustical equivalent circuit
Electromechanicoacoustical systems such as loudspeakers can be modeled as equivalent electrical circuits as long as each element moves as a whole. This is usually the case at low frequencies or at frequencies where the dimensions of the system are small compared to the wavelength of interest. To obtain a complete model of the loudspeaker, the interactions and properties of electrical, mechanical, and acoustical subsystems composing the loudspeaker driver must each be modeled. The following sections detail how the circuit may be obtained starting with the amplifier and ending with the acoustical load presented by air. A similar development can be found in [1] or [2].
Electrical subsystem
The electrical part of the system is composed of a driving amplifier and a voice coil. Most amplifiers can be approximated as a perfect voltage source in series with the amplifier output impedance. The voice coil exhibits an inductance and a resistance that may be directly modeled as a circuit.
Electrical to mechanical subsystem
When the loudspeaker is fed an electrical signal, the voice coil and magnet convert current to force. Similarly, voltage is related to the velocity. This relationship between the electrical side and the mechanical side can be modeled by a transformer.
;
Mechanical subsystem
In a first approximation, a moving coil loudspeaker may be thought of as a massspring system where the diaphragm and the voice coil constitute the mass and the spider and surround constitute the spring element. Losses in the suspension can be modeled as a resistor.
The equation of motion gives us :
Which yields the mechanical impedance type analogy in the form of a series RLC circuit. A parallel RLC circuit may also be obtained to get the mobility analog following mathematical manipulation:
Which expresses the mechanical mobility type analogy in the form of a parallel RLC circuit where the denominator elements are respectively a parallel conductance, inductance, and compliance.
Mechanical to acoustical subsystem
A loudspeaker’s diaphragm may be thought of as a piston that pushes and pulls on the air facing it, converting mechanical force and velocity into acoustic pressure and volume velocity. The equations are as follow:
;
These equations can be modeled by a transformer.
Acoustical subsystem
The impedance presented by the air load on the loudspeaker's diaphragm is both resistive due to sound radiation and reactive due to the air mass that is being pushed radially but does not contribute to sound radiation to the far field. The air load on the diaphragm can be modeled as an impedance or an admittance. Specific values and approximations can be found in [1], [2] or [3]. Note that the air load depends on the mounting conditions of the loudspeaker. If the loudspeaker is mounted in a baffle, the air load will be the same on each side of the diaphragm. Then, if the air load on one side is in the admittance analogy, then the total air load is as both loads are in parallel.
Complete electromechanoacoustical equivalent circuit
Using electrical impedance, mechanical mobility and acoustical admittance yield the following equivalent circuit, modeling the entire loudspeaker drive unit.
This circuit can be reduced by substituting the transformers and connected loads by an equivalent loading that would present the same impedance as the loaded transformer. An example of this is shown on figure 7, where acoustical and electrical loads and sources have been "brought over" to the mechanical side.
The advantage of doing such manipulations is that we can then directly relate electrical measurements with elements in the circuit. This will later allow us to obtain values for the different components of the model and match this model to real loudspeaker drivers. We can further simplify this circuit by using Norton's theorem and converting the series electrical components and voltage source into an equivalent current source and parallel electrical components. Then, using a technique called the Dot method, presented in section Solution Methods: ElectroMechanical Analogies, we can obtain a single loop series circuit which is the dual of the parallel circuit previously obtained with Norton's theorem. If we are mainly interested in the low frequency behavior of the loudspeaker, as should be the case when using lumped element modeling, we can neglect the effect of the voice coil inductance, which has an effect only at high frequencies. Furthermore, the air load impedance at low frequencies is masslike and can be modeled by a simple inductance . This results in a simplified low frequency model equivalent circuit, shown of figure 8, which is easier to manipulate than the circuit of figure 7. Note that the analogy used for this circuit is of the impedance type.
Where if is the radius of the loudspeaker and , the density of air. Mass elements, in this case the mass of the diaphragm and voice coil and the air mass loading the diaphragm can be regrouped in a single element:
ThieleSmall Parameters
Theory
The complete low frequency behavior of a loudspeaker drive unit can be modeled with just six parameters, called ThieleSmall parameters. Most of these parameters result from algebraic manipulation of the equations of the circuit of figure 8. Loudspeaker driver manufacturers seldom provide electromechanoacoustical parameters directly and rather provide ThieleSmall parameters in datasheets, but conversion from one to the other is quite simple. The ThieleSmall parameters are as follow:
1. , the voice coil DC resistance;
2. , the electrical Q factor;
3. , the mechanical Q factor;
4. , the loudspeaker resonance frequency;
5. , the effective surface area of the diaphragm;
6. , the equivalent suspension volume: the volume of air that has the same acoustic compliance as the suspension of the loudspeaker driver.
These parameters can be related directly from the low frequency approximation circuit of figure 8, with and being explicit.
; ; ;
Where is the Bulk modulus of air. It follows that, if given ThieleSmall parameters, one can extract the values of each component of the circuit of figure 8 using the following equations :
; ; ; ; ;
Measurement
Many methods can be used to measure ThieleSmall parameters of drivers. Measurement of ThieleSmall parameters is sometimes necessary if a manufacturer does not provide them. Also, the actual ThieleSmall parameters of a given loudspeaker can differ from nominal values significantly. The method described in this section comes from [2]. Note that for this method, the loudspeaker is considered to be mounted in an infinite baffle. In practice, a baffle with a diameter of four times that of the loudspeaker is sufficient. Measurements without a baffle are also possible: the air mass loading will simply be halved and can be easily accounted for. The setup for this method includes an FFT analyzer or a mean to obtain an impedance curve. A signal generator of variable frequency and an AC meter can also be used.
Once the impedance curve of the loudspeaker is measured, and can be directly identified by looking at the low frequency asymptote of the impedance value and the center frequency of the resonance peak. If the frequencies where are identified as and , Q factors can be calculated.
can simply be approximated by , where is the radius of the loudspeaker driver. The last remaining ThieleSmall parameter, is slightly trickier to measure. The idea is to either increase mass or reduce compliance of the loudspeaker drive unit and note the shift in resonance frequency. If a known mass is added to the loudspeaker diaphragm, the new resonance frequency will be:
And the equivalent suspension volume may be obtained with:
Hence, all ThieleSmall parameters modeling the low frequency behavior of the loudspeaker drive unit can be obtained from a fairly simple setup. These parameters are of tremendous help in loudspeaker enclosure design.
Numerical example
This section presents a numerical example of obtaining ThieleSmall parameters from impedance curves. The impedance curves presented in this section have been obtained from simulations using nominal ThieleSmall parameters of a real woofer loudspeaker. Firsy, these ThieleSmall parameters have been transformed into an electromechanoacoustical circuit using the equation presented before. Second, the circuit was treated as a black box and the method to extract ThieleSmall parameters was used. The purpose of this simulation is to present the method, step by step, using realistic values so that the reader can get more familiar with the process, the magnitude of the values and with what to expect when performing such measurements.
For this simulation, a loudspeaker of radius is mounted on a baffle sufficiently large to act as an infinite baffle. Its impedance is obtained and plotted in figure 11, where important cursors have already been placed.
The low frequency asymptote is immediately identified as . The resonance is clear and centered at . The value of the impedance at this frequency is about . This yields , which occurs at and . With this information, we can compute some of the ThieleSmall parameters.
As a next step, a mass of is fixed to the loudspeaker diaphragm. This shifts the resonance frequency and yields a new impedance curve, as shown on figure 12.
Once all six ThieleSmall parameters have been obtained, it is possible to calculate values for the electromechanoacoustical circuit modeling elements of figure 6 or 7. From then, the design of an enclosure can start. This is discussed in application sections Sealed box subwoofer design and Bass reflex enclosure design.
References
[1] Kleiner, Mendel. Electroacoustics. CRC Press, 2013.
[2] Beranek, Leo L., and Tim Mellow. Acoustics: sound fields and transducers. Academic Press, 2012.
[3] Kinsler, Lawrence E., et al. Fundamentals of Acoustics, 4th Edition. WileyVCH, 1999.
[4] Small, Richard H. "Direct radiator loudspeaker system analysis." Journal of the Audio Engineering Society 20.5 (1972): 383395.
Moving Resonators
Moving Resonators
Consider the situation shown in the figure below. We have a typical Helmholtz resonator driven by a massless piston which generates a sinusoidal pressure , however the cavity is not fixed in this case. Rather, it is supported above the ground by a spring with compliance . Assume the cavity has a mass .
Recall the Helmholtz resonator (see Module #9). The difference in this case is that the pressure in the cavity exerts a force on the bottom of the cavity, which is now not fixed as in the original Helmholtz resonator. This pressure causes a force that acts upon the cavity bottom. If the surface area of the cavity bottom is , then Newton's Laws applied to the cavity bottom give
In order to develop the equivalent circuit, we observe that we simply need to use the pressure (potential across ) in the cavity to generate a force in the mechanical circuit. The above equation shows that the mass of the cavity and the spring compliance should be placed in series in the mechanical circuit. In order to convert the pressure to a force, the transformer is used with a ratio of .
Example
A practical example of a moving resonator is a marimba. A marimba is a similar to a xylophone but has larger resonators that produce deeper and richer tones. The resonators (seen in the picture as long, hollow pipes) are mounted under an array of wooden bars which are struck to create tones. Since these resonators are not fixed, but are connected to the ground through a stiffness (the stand), it can be modeled as a moving resonator. Marimbas are not tunable instruments like flutes or even pianos. It would be interesting to see how the tone of the marimba changes as a result of changing the stiffness of the mount.
For more information about the acoustics of marimbas see http://www.mostlymarimba.com/techno1.html
Part 2: OneDimensional Wave Motion
Transverse vibrations of strings
Introduction
This section deals with the wave nature of vibrations constrained to one dimension. Examples of this type of wave motion are found in objects such a pipes and tubes with a small diameter (no transverse motion of fluid) or in a string stretched on a musical instrument.
Stretched strings can be used to produce sound (e.g. music instruments like guitars). The stretched string constitutes a mechanical system that will be studied in this chapter. Later, the characteristics of this system will be used to help to understand by analogies acoustical systems.
What is a wave equation?
There are various types of waves (i.e. electromagnetic, mechanical, etc.) that act all around us. It is important to use wave equations to describe the timespace behavior of the variables of interest in such waves. Wave equations solve the fundamental equations of motion in a way that eliminates all variables but one. Waves can propagate longitudinal or parallel to the propagation direction or perpendicular (transverse) to the direction of propagation. To visualize the motion of such waves click here (Acoustics animations provided by Dr. Dan Russell,Kettering University)
One dimensional Case
Assumptions :
 the string is uniform in size and density
 stiffness of string is negligible for small deformations
 effects of gravity neglected
 no dissipative forces like frictions
 string deforms in a plane
 motion of the string can be described by using one single spatial coordinate
Spatial representation of the string in vibration:
The following is the freebody diagram of a string in motion in a spatial coordinate system:
From the diagram above, it can be observed that the tensions in each side of the string will be the same as follows:
Using Taylor series to expand we obtain:
Characterization of the mechanical system
A one dimensional wave can be described by the following equation (called the wave equation):
where,
is a solution,
With and
This is the D'Alambert solution, for more information see: [1]
Another way to solve this equation is the Method of separation of variables. This is useful for modal analysis. This assumes the solution is of the form:
The result is the same as above, but in a form that is more convenient for modal analysis.
For more information on this approach see: Eric W. Weisstein et al. "Separation of Variables." From MathWorldA Wolfram Web Resource. [2]
Please see Wave Properties for information on variable c, along with other important properties.
For more information on wave equations see: Eric W. Weisstein. "Wave Equation." From MathWorldA Wolfram Web Resource. [3]
Example with the function :
Example: Java String simulation
This show a simple simulation of a plucked string with fixed ends.
TimeDomain Solutions
d'Alembert Solutions
In 1747, Jean Le Rond d'Alembertpublished a solution to the onedimensional wave equation.
The general solution, now known as the d'Alembert method, can be found by introducing two new variables:
and
and then applying the chain rule to the general form of the wave equation.
From this, the solution can be written in the form:
where f and g are arbitrary functions, that represent two waves traveling in opposing directions.
A more detailed look into the proof of the d'Alembert solution can be found here.
Example of Time Domain Solution
If f(ctx) is plotted vs. x for two instants in time, the two waves are the same shape but the second displaced by a distance of c(t2t1) to the right.
The two arbitrary functions could be determined from initial conditions or boundary values.
Boundary Conditions and Forced Vibrations
Boundary Conditions
The functions representing the solutions to the wave equation previously discussed,
i.e. with and
are dependent upon the boundary and initial conditions. If it is assumed that the wave is propagating through a string, the initial conditions are related to the specific disturbance in the string at t=0. These specific disturbances are determined by location and type of contact and can be anything from simple oscillations to violent impulses. The effects of boundary conditions are less subtle.
The most simple boundary conditions are the Fixed Support and Free End. In practice, the Free End boundary condition is rarely encountered since it is assumed there are no transverse forces holding the string (e.g. the string is simply floating).
For a Fixed Support:
The overall displacement of the waves travelling in the string, at the support, must be zero. Denoting x=0 at the support, This requires:
Therefore, the total transverse displacement at x=0 is zero.
The sequence of wave reflection for incident, reflected and combined waves are illustrated below. Please note that the wave is traveling to the left (negative x direction) at the beginning. The reflected wave is ,of course, traveling to the right (positive x direction).
For a Free Support:
Unlike the Fixed Support boundary condition, the transverse displacement at the support does not need to be zero, but must require the sum of transverse forces to cancel. If it is assumed that the angle of displacement is small,
and so,
But of course, the tension in the string, or T, will not be zero and this requires the slope at x=0 to be zero:
i.e.
Again for free boundary, the sequence of wave reflection for incident, reflected and combined waves are illustrated below:
Other Boundary Conditions:
There are many other types of boundary conditions that do not fall into our simplified categories. As one would expect though, it isn't difficult to relate the characteristics of numerous "complex" systems to the basic boundary conditions. Typical or realistic boundary conditions include massloaded, resistanceloaded, dampingloaded, and impedanceloaded strings. For further information, see Kinsler, Fundamentals of Acoustics, pp 5458.
Here is a website with nice movies of wave reflection at different BC's: Wave Reflection
Wave Properties
To begin with, a few definitions of useful variables will be discussed. These include; the wave number, phase speed, and wavelength characteristics of wave travelling through a string.
The speed that a wave propagates through a string is given in terms of the phase speed, typically in m/s, given by:
where is the density per unit length of the string.
The wave number is used to reduce the transverse displacement equation to a simpler form and for simple harmonic motion, is multiplied by the lateral position. It is given by:
where
Lastly, the wavelength is defined as:
and is defined as the distance between two points, usually peaks, of a periodic waveform.
These "wave properties" are of practical importance when calculating the solution of the wave equation for a number of different cases. As will be seen later, the wave number is used extensively to describe wave phenomenon graphically and quantitatively.
For further information: Wave Properties
Forced Vibrations
1.forced vibrations of infinite string suppose there is a string very long , at x=0 there is a force exerted on it.
F(t)=Fcos(wt)=Real{Fexp(jwt)}
use the boundary condition at x=0,
neglect the reflected wave
it is easy to get the wave form
where w is the angular velocity, k is the wave number.
according to the impedance definition
it represents the characteristic impedance of the string. obviously, it is purely resistive, which is like the resistance in the mechanical system.
The dissipated power
Note: along the string, all the variables propagate at same speed.
link title a useful link to show the timespace property of the wave.
Some interesting animations of the wave at different boundary conditions.
1.hard boundary (which is like a fixed end)
2.soft boundary (which is like a free end)
3.from low density to high density string
4.from high density to low density string
Part 3: Applications
Room Acoustics and Concert Halls
Room Acoustics and Concert Halls
Introduction
From performing on many different rooms and stages all over the United States, I thought it would be nice to have a better understanding and source about the room acoustics. This Wikibook page is intended to help the user with basic technical questions/answers about room acoustics. Main topics that will be covered are: what really makes a room sound good or bad, alive or dead. This will lead into absorption and transmission coefficients, decay of sound in the room, and reverberation. Different use of materials in rooms will be mentioned also. There is no intention of taking work from another. This page is a switchboard source to help the user find information about room acoustics.
Sound Fields
Two types of sound fields are involved in room acoustics: Direct Sound and Reverberant Sound.
Direct Sound
The component of the sound field in a room that involves only a direct path between the source and the receiver, before any reflections off walls and other surfaces.
Reverberant Sound
The component of the sound field in a room that involves the direct path and the path after it reflects off of walls or any other surfaces. How the waves deflect off of the mediums all depends on the absorption and transmission coefficients.
Good example pictures are shown at Crutchfield Advisor, a Physics Site from MTSU, and Voiceteacher.com
Room Coefficients
In a perfect world, if there is a sound shot right at a wall, the sound should come right back. But because sounds hit different materials types of walls, the sound does not have perfect reflection. From 1, these are explained as follows:
Absorption & Transmission Coefficients
The best way to explain how sound reacts to different mediums is through acoustical energy. When sound impacts on a wall, acoustical energy will be reflected, absorbed, or transmitted through the wall.
If all of the acoustic energy hits the wall and none is reflected, the alpha would equal 1. The energy had zero reflection and was absorbed or transmitted. This would be an example of a dead or soft wall because it takes in everything and doesn't reflect anything back. Rooms that are like this are called Anechoic Rooms which looks like this from Axiomaudio.
If all of the acoustic energy hits the wall and all reflects back, the alpha would equal 0. This would be an example of a live or hard wall because the sound bounces right back and does not go through the wall. Rooms that are like this are called Reverberant Rooms like this McIntoshroom. Look how the walls have nothing attached to them. More room for the sound waves to bounce off the walls.
Room Averaged Sound Absorption Coefficient
Not all rooms have the same walls on all sides. The room averaged sound absorption coefficient can be used to have different types of materials and areas of walls averaged together.
Absorption Coefficients for Specific Materials
Basic sound absorption Coefficients are shown here at Acoustical Surfaces.
Brick, unglazed, painted alpha ~ .01  .03 > Sound reflects back
An open door alpha equals 1 > Sound goes through
Units are in Sabins.
Sound Decay and Reverberation Time
In a large reverberant room, a sound can still propagate after the sound source has been turned off. This time when the sound intensity level has decay 60 dB is called the reverberation time of the room.
Great Halls in the World
Pick Staiger at Northwestern U
References
[1] Lord, Gatley, Evensen; Noise Control for Engineers, Krieger Publishing, 435 pgs
Back to Engineering Acoustics
Created by Kevin Baldwin
Bass Reflex Enclosure Design
Introduction
Bassreflex enclosures improve the lowfrequency response of loudspeaker systems. Bassreflex enclosures are also called "ventedbox design" or "portedcabinet design". A bassreflex enclosure includes a vent or port between the cabinet and the ambient environment. This type of design, as one may observe by looking at contemporary loudspeaker products, is still widely used today. Although the construction of bassreflex enclosures is fairly simple, their design is not simple, and requires proper tuning. This reference focuses on the technical details of bassreflex design. General loudspeaker information can be found here.
Effects of the Port on the Enclosure Response
Before discussing the bassreflex enclosure, it is important to be familiar with the simpler sealed enclosure system performance. As the name suggests, the sealed enclosure system attaches the loudspeaker to a sealed enclosure (except for a small air leak included to equalize the ambient pressure inside). Ideally, the enclosure would act as an acoustical compiance element, as the air inside the enclosure is compressed and rarified. Often, however, an acoustic material is added inside the box to reduce standing waves, dissipate heat, and other reasons. This adds a resistive element to the acoustical lumpedelement model. A nonideal model of the effect of the enclosure actually adds an acoustical mass element to complete a series lumpedelement circuit given in Figure 1. For more on sealed enclosure design, see the Sealed Box Subwoofer Design page.
Figure 1. Sealed enclosure acoustic circuit.
In the case of a bassreflex enclosure, a port is added to the construction. Typically, the port is cylindrical and is flanged on the end pointing outside the enclosure. In a bassreflex enclosure, the amount of acoustic material used is usually much less than in the sealed enclosure case, often none at all. This allows air to flow freely through the port. Instead, the larger losses come from the air leakage in the enclosure. With this setup, a lumpedelement acoustical circuit has the following form.
In this figure, represents the radiation impedance of the outside environment on the loudspeaker diaphragm. The loading on the rear of the diaphragm has changed when compared to the sealed enclosure case. If one visualizes the movement of air within the enclosure, some of the air is compressed and rarified by the compliance of the enclosure, some leaks out of the enclosure, and some flows out of the port. This explains the parallel combination of , , and . A truly realistic model would incorporate a radiation impedance of the port in series with , but for now it is ignored. Finally, , the acoustical mass of the enclosure, is included as discussed in the sealed enclosure case. The formulas which calculate the enclosure parameters are listed in Appendix B.
It is important to note the parallel combination of and . This forms a Helmholtz resonator (click here for more information). Physically, the port functions as the “neck” of the resonator and the enclosure functions as the “cavity.” In this case, the resonator is driven from the piston directly on the cavity instead of the typical Helmholtz case where it is driven at the “neck.” However, the same resonant behavior still occurs at the enclosure resonance frequency, . At this frequency, the impedance seen by the loudspeaker diaphragm is large (see Figure 3 below). Thus, the load on the loudspeaker reduces the velocity flowing through its mechanical parameters, causing an antiresonance condition where the displacement of the diaphragm is a minimum. Instead, the majority of the volume velocity is actually emitted by the port itself instead of the loudspeaker. When this impedance is reflected to the electrical circuit, it is proportional to , thus a minimum in the impedance seen by the voice coil is small. Figure 3 shows a plot of the impedance seen at the terminals of the loudspeaker. In this example, was found to be about 40 Hz, which corresponds to the null in the voicecoil impedance.
Figure 3. Impedances seen by the loudspeaker diaphragm and voice coil.
Quantitative Analysis of Port on Enclosure
The performance of the loudspeaker is first measured by its velocity response, which can be found directly from the equivalent circuit of the system. As the goal of most loudspeaker designs is to improve the bass response (leaving highfrequency production to a tweeter), low frequency approximations will be made as much as possible to simplify the analysis. First, the inductance of the voice coil, , can be ignored as long as . In a typical loudspeaker, is of the order of 1 mH, while is typically 8, thus an upper frequency limit is approximately 1 kHz for this approximation, which is certainly high enough for the frequency range of interest.
Another approximation involves the radiation impedance, . It can be shown [1] that this value is given by the following equation (in acoustical ohms):
Where and are types of Bessel functions. For small values of ka,
and 
Hence, the lowfrequency impedance on the loudspeaker is represented with an acoustic mass [1]. For a simple analysis, , , , and (the transducer parameters, or ThieleSmall parameters) are converted to their acoustical equivalents. All conversions for all parameters are given in Appendix A. Then, the series masses, , , and , are lumped together to create . This new circuit is shown below.
Unlike sealed enclosure analysis, there are multiple sources of volume velocity that radiate to the outside environment. Hence, the diaphragm volume velocity, , is not analyzed but rather . This essentially draws a “bubble” around the enclosure and treats the system as a source with volume velocity . This “lumped” approach will only be valid for low frequencies, but previous approximations have already limited the analysis to such frequencies anyway. It can be seen from the circuit that the volume velocity flowing into the enclosure, , compresses the air inside the enclosure. Thus, the circuit model of Figure 3 is valid and the relationship relating input voltage, to may be computed.
In order to make the equations easier to understand, several parameters are combined to form other parameter names. First, and , the enclosure and loudspeaker resonance frequencies, respectively, are:
Based on the nature of the derivation, it is convenient to define the parameters and h, the Helmholtz tuning ratio:
A parameter known as the compliance ratio or volume ratio, , is given by:
Other parameters are combined to form what are known as quality factors:
This notation allows for a simpler expression for the resulting transfer function [1]:
where
Development of LowFrequency Pressure Response
It can be shown [2] that for , a loudspeaker behaves as a spherical source. Here, a represents the radius of the loudspeaker. For a 15” diameter loudspeaker in air, this low frequency limit is about 150 Hz. For smaller loudspeakers, this limit increases. This limit dominates the limit which ignores , and is consistent with the limit that models by .
Within this limit, the loudspeaker emits a volume velocity , as determined in the previous section. For a simple spherical source with volume velocity , the farfield pressure is given by [1]:
It is possible to simply let for this analysis without loss of generality because distance is only a function of the surroundings, not the loudspeaker. Also, because the transfer function magnitude is of primary interest, the exponential term, which has a unity magnitude, is omitted. Hence, the pressure response of the system is given by [1]:
Where . In the following sections, design methods will focus on rather than , which is given by:
This also implicitly ignores the constants in front of since they simply scale the response and do not affect the shape of the frequency response curve.
Alignments
A popular way to determine the ideal parameters has been through the use of alignments. The concept of alignments is based upon well investigated electrical filter theory. Filter development is a method of selecting the poles (and possibly zeros) of a transfer function to meet a particular design criterion. The criteria are the desired properties of a magnitudesquared transfer function, which in this case is . From any of the design criteria, the poles (and possibly zeros) of are found, which can then be used to calculate the numerator and denominator. This is the “optimal” transfer function, which has coefficients that are matched to the parameters of to compute the appropriate values that will yield a design that meets the criteria.
There are many different types of filter designs, each which have tradeoffs associated with them. However, this design approach is limited because of the structure of . In particular, it has the structure of a fourthorder highpass filter with all zeros at s = 0. Therefore, only those filter design methods which produce a lowpass filter with only poles will be acceptable methods to use. From the traditional set of algorithms, only Butterworth and Chebyshev lowpass filters have only poles. In addition, another type of filter called a quasiButterworth filter can also be used, which has similar properties to a Butterworth filter. These three algorithms are fairly simple, thus they are the most popular. When these lowpass filters are converted to highpass filters, the transformation produces in the numerator.
More details regarding filter theory and these relationships can be found in numerous resources, including [5].
Butterworth Alignment
The Butterworth algorithm is designed to have a maximally flat pass band. Since the slope of a function corresponds to its derivatives, a flat function will have derivatives equal to zero. Since as flat of a pass band as possible is optimal, the ideal function will have as many derivatives equal to zero as possible at s = 0. Of course, if all derivatives were equal to zero, then the function would be a constant, which performs no filtering.
Often, it is better to examine what is called the loss function. Loss is the reciprocal of gain, thus
The loss function can be used to achieve the desired properties, then the desired gain function is recovered from the loss function.
Now, applying the desired Butterworth property of maximal passband flatness, the loss function is simply a polynomial with derivatives equal to zero at s = 0. At the same time, the original polynomial must be of degree eight (yielding a fourthorder function). However, derivatives one through seven can be equal to zero if [3]
With the highpass transformation ,
It is convenient to define , since or 3 dB. This definition allows the matching of coefficients for the describing the loudspeaker response when . From this matching, the following design equations are obtained [1]:
QuasiButterworth Alignment
The quasiButterworth alignments do not have as welldefined of an algorithm when compared to the Butterworth alignment. The name “quasiButterworth” comes from the fact that the transfer functions for these responses appear similar to the Butterworth ones, with (in general) the addition of terms in the denominator. This will be illustrated below. While there are many types of quasiButterworth alignments, the simplest and most popular is the 3rd order alignment (QB3). The comparison of the QB3 magnitudesquared response against the 4th order Butterworth is shown below.
Notice that the case is the Butterworth alignment. The reason that this QB alignment is called 3rd order is due to the fact that as B increases, the slope approaches 3 dec/dec instead of 4 dec/dec, as in 4th order Butterworth. This phenomenon can be seen in Figure 5.
Equating the system response with , the equations guiding the design can be found [1]:
Chebyshev Alignment
The Chebyshev algorithm is an alternative to the Butterworth algorithm. For the Chebyshev response, the maximallyflat passband restriction is abandoned. Now, a ripple, or fluctuation is allowed in the pass band. This allows a steeper transition or rolloff to occur. In this type of application, the lowfrequency response of the loudspeaker can be extended beyond what can be achieved by Butterworthtype filters. An example plot of a Chebyshev highpass response with 0.5 dB of ripple against a Butterworth highpass response for the same is shown below.
The Chebyshev response is defined by [4]:
is called the Chebyshev polynomial and is defined by [4]:
Fortunately, Chebyshev polynomials satisfy a simple recursion formula [4]:
For more information on Chebyshev polynomials, see the Wolfram Mathworld: Chebyshev Polynomials page.
When applying the highpass transformation to the 4th order form of , the desired response has the form [1]:
The parameter determines the ripple. In particular, the magnitude of the ripple is dB and can be chosen by the designer, similar to B in the quasiButterworth case. Using the recursion formula for ,
Applying this equation to [1],
Thus, the design equations become [1]:
Choosing the Correct Alignment
With all the equations that have already been presented, the question naturally arises, “Which one should I choose?” Notice that the coefficients , , and are not simply related to the parameters of the system response. Certain combinations of parameters may indeed invalidate one or more of the alignments because they cannot realize the necessary coefficients. With this in mind, general guidelines have been developed to guide the selection of the appropriate alignment. This is very useful if one is designing an enclosure to suit a particular transducer that cannot be changed.
The general guideline for the Butterworth alignment focuses on and . Since the three coefficients , , and are a function of , , h, and , fixing one of these parameters yields three equations that uniquely determine the other three. In the case where a particular transducer is already given, is essentially fixed. If the desired parameters of the enclosure are already known, then is a better starting point.
In the case that the rigid requirements of the Butterworth alignment cannot be satisfied, the quasiButterworth alignment is often applied when is not large enough.. The addition of another parameter, B, allows more flexibility in the design.
For values that are too large for the Butterworth alignment, the Chebyshev alignment is typically chosen. However, the steep transition of the Chebyshev alignment may also be utilized to attempt to extend the bass response of the loudspeaker in the case where the transducer properties can be changed.
In addition to these three popular alignments, research continues in the area of developing new algorithms that can manipulate the lowfrequency response of the bassreflex enclosure. For example, a 5th order quasiButterworth alignment has been developed [6]; its advantages include improved low frequency extension, and much reduced driver excursion at low frequencies and typically biamping or triamping, while its disadvatages include somewhat difficult mathematics and electronic complication (electronic crossovers are typically required). Another example [7] applies rootlocus techniques to achieve results. In the modern age of highpowered computing, other researchers have focused their efforts in creating computerized optimization algorithms that can be modified to achieve a flatter response with sharp rolloff or introduce quasiripples which provide a boost in subbass frequencies [8].
References
[1] Leach, W. Marshall, Jr. Introduction to Electroacoustics and Audio Amplifier Design. 2nd ed. Kendall/Hunt, Dubuque, IA. 2001.
[2] Beranek, L. L. Acoustics. 2nd ed. Acoustical Society of America, Woodbridge, NY. 1993.
[3] DeCarlo, Raymond A. “The Butterworth Approximation.” Notes from ECE 445. Purdue University. 2004.
[4] DeCarlo, Raymond A. “The Chebyshev Approximation.” Notes from ECE 445. Purdue University. 2004.
[5] VanValkenburg, M. E. Analog Filter Design. Holt, Rinehart and Winston, Inc. Chicago, IL. 1982.
[6] Kreutz, Joseph and Panzer, Joerg. "Derivation of the QuasiButterworth 5 Alignments." Journal of the Audio Engineering Society. Vol. 42, No. 5, May 1994.
[7] Rutt, Thomas E. "RootLocus Technique for VentedBox Loudspeaker Design." Journal of the Audio Engineering Society. Vol. 33, No. 9, September 1985.
[8] Simeonov, Lubomir B. and ShopovaSimeonova, Elena. "PassiveRadiator Loudspeaker System Design Software Including Optimization Algorithm." Journal of the Audio Engineering Society. Vol. 47, No. 4, April 1999.
Appendix A: Equivalent Circuit Parameters
Name  Electrical Equivalent  Mechanical Equivalent  Acoustical Equivalent 

VoiceCoil Resistance  
Driver (Speaker) Mass  See  
Driver (Speaker) Suspension Compliance  
Driver (Speaker) Suspension Resistance  
Enclosure Compliance  
Enclosure AirLeak Losses  
Acoustic Mass of Port  
Enclosure Mass Load  See  See  
LowFrequency Radiation Mass Load  See  See  
Combination Mass Load 
Appendix B: Enclosure Parameter Formulas
Based on these dimensions [1],
(inside enclosure gross volume)  (baffle area of the side the speaker is mounted on) 
specific heat of air at constant isovolumetric process (about at 300 K)  specific heat of filling at constant volume () 
mean density of air (about at 300 K)  density of filling 
ratio of specific heats (Isobaric/Isovolumetric processes) for air (about 1.4 at 300 K)  speed of sound in air (about 344 m/s) 
= effective density of enclosure. If little or no filling (acceptable assumption in a bassreflex system but not for sealed enclosures), 
New Acoustic Filter For Ultrasonics Media
Introduction
Acoustic filters are used in many devices such as mufflers, noise control materials (absorptive and reactive), and loudspeaker systems to name a few. Although the waves in simple (singlemedium) acoustic filters usually travel in gases such as air and carbonmonoxide (in the case of automobile mufflers) or in materials such as fiberglass, polyvinylidene fluoride (PVDF) film, or polyethylene (Saran Wrap), there are also filters that couple two or three distinct media together to achieve a desired acoustic response. General information about basic acoustic filter design can be perused at the following wikibook page [Acoustic Filter Design & Implementation]. The focus of this article will be on acoustic filters that use multilayer air/polymer filmcoupled media as its acoustic medium for sound waves to propagate through; concluding with an example of how these filters can be used to detect and extrapolate audio frequency information in highfrequency "carrier" waves that carry an audio signal. However, before getting into these specific type of acoustic filters, we need to briefly discuss how sound waves interact with the medium(media) in which it travels and how these factors can play a role when designing acoustic filters.
Changes in Media Properties Due to Sound Wave Characteristics
As with any system being designed, the filter response characteristics of an acoustic filter are tailored based on the frequency spectrum of the input signal and the desired output. The input signal may be infrasonic (frequencies below human hearing), sonic (frequencies within human hearing range), or ultrasonic (frequencies above human hearing range). In addition to the frequency content of the input signal, the density, and, thus, the characteristic impedance of the medium (media) being used in the acoustic filter must also be taken into account. In general, the characteristic impedance for a particular medium is expressed as...= (equilibrium) density of medium = speed of sound in medium
is intrepreted as the (timeaveraged) rate of energy transmission of a sound wave through a unit area normal to the direction of propagation, and this parameter is also an important factor in acoustic filter design because the characteristic properties of the given medium can change relative to intensity of the sound wave traveling through it. In other words, the reaction of the particles (atoms or molecules) that make up the medium will respond differently when the intensity of the sound wave is very high or very small relative to the size of the control area (i.e. dimensions of the filter, in this case). Other properties such as the elasticity and mean propagation velocity (of a sound wave) can change in the acoustic medium as well, but focusing on frequency, impedance, and/or intensity in the design process usually takes care of these other parameters because most of them will inevitably be dependent on the aforementioned properties of the medium.
Why Coupled Acoustic Media in Acoustic Filters?
In acoustic transducers, media coupling is employed in acoustic transducers to either increase or decrease the impedance of the transducer, and, thus, control the intensity and speed of the signal acting on the transducer while converting the incident wave, or initial excitation sound wave, from one form of energy to another (e.g. converting acoustic energy to electrical energy). Specifically, the impedance of the transducer is augmented by inserting a solid structure (not necessarily rigid) between the transducer and the initial propagation medium (e.g. air). The reflective properties of the inserted medium is exploited to either increase or decrease the intensity and propagation speed of the incident sound wave. It is the ability to alter, and to some extent, control, the impedance of a propagation medium by (periodically) inserting (a) solid structure(s) such as thin, flexible films in the original medium (air) and its ability to concomitantly alter the frequency response of the original medium that makes use of multilayer media in acoustic filters attractive. The reflection factor and transmission factor and , respectively, between two media, expressed as...
are the tangible values that tell how much of the incident wave is being reflected from and transmitted through the junction where the media meet. Note that is the (total) input impedance seen by the incident sound wave upon just entering an airsolid acoustic media layer. In the case of multiple aircolumns as shown in Fig. 2, is the aggregate impedance of each aircolumn layer seen by the incident wave at the input. Below in Fig. 1, a simple illustration explains what happens when an incident sound wave propagating in medium (1) and comes in contact with medium (2) at the junction of the both media (x=0), where the sound waves are represented by vectors.
As mentioned above, an example of three such successive airsolid acoustic media layers is shown in Fig. 2 and the electroacoustic equivalent circuit for Fig. 2 is shown in Fig. 3 where = (density of solid material)(thickness of solid material) = unitarea (or volume) mass, characteristic acoustic impedance of medium, and wavenumber. Note that in the case of a multilayer, coupled acoustic medium in an acoustic filter, the impedance of each airsolid section is calculated by using the following general purpose impedance ratio equation (also referred to as transfer matrices)...
where is the (known) impedance at the edge of the solid of an airsolid layer (on the right) and is the (unknown) impedance at the edge of the air column of an airsolid layer.
Effects of HighIntensity, Ultrasonic Waves in Acoustic Media in Audio Frequency Spectrum
When an ultrasonic wave is used as a carrier to transmit audio frequencies, three audio effects are associated with extrapolating the audio frequency information from the carrier wave: (a) beating effects, (b) parametric array effects, and (c) radiation pressure.
Beating occurs when two ultrasonic waves with distinct frequencies and propagate in the same direction, resulting in amplitude variations which consequently make the audio signal information go in and out of phase, or “beat”, at a frequency of .
Parametric array effects occur when the intensity of an ultrasonic wave is so high in a particular medium that the high displacements of particles (atoms) per wave cycle changes properties of that medium so that it influences parameters like elasticity, density, propagation velocity, etc. in a nonlinear fashion. The results of parametric array effects on modulated, highintensity, ultrasonic waves in a particular medium (or coupled media) is the generation and propagation of audio frequency waves (not necessarily present in the original audio information) that are generated in a manner similar to the nonlinear process of amplitude demodulation commonly inherent in diode circuits (when diodes are forward biased).
Another audio effect that arises from highintensity ultrasonic beams of sound is a static (DC) pressure called radiation pressure. Radiation pressure is similar to parametric array effects in that amplitude variations in the signal give rise to audible frequencies via amplitude demodulation. However, unlike parametric array effects, radiation pressure fluctuations that generate audible signals from amplitude demodulation can occur due to any lowfrequency modulation and not just from pressure fluctuations occurring at the modulation frequency or beating frequency .
An Application of Coupled Media in Acoustic Filters
Figs. 1  3 were all from a research paper entitled New Type of Acoustics Filter Using Periodic Polymer Layers for Measuring Audio Signal Components Excited by AmplitudeModulated High_Intensity Ultrasonic Wavessubmitted to the Audio Engineering Society (AES) by Minoru Todo, Primary Innovator at Measurement Specialties, Inc., in the October 2005 edition of the AES Journal. Figs. 4 and 5 below, also from this paper, are illustrations of test setups referred to in this paper. Specifically, Fig. 4 is a test setup used to measure the transmission (of an incident ultrasonic sound wave) through the acoustic filter described by Figs. 1 and 2. Fig. 5 is a block diagram of the test setup used for measuring radiation pressure, one of the audio effects mentioned in the previous section. It turns out that out of all of the audio effects mentioned in the previous section that are caused by highintensity ultrasonic waves propagating in a medium, sound waves produced from radiated pressure are the hardest to detect when microphones and preamplifiers are used in the detection/receiver system. Although nonlinear noise artifacts occur due to overloading of the preamplifier present in the detection/receiver system, the bulk of the nonlinear noise comes from the inherent nonlinear noise properties of microphones. This is true because all microphones, even specialized measurement microphones designed for audio spectrum measurements that have sensitivity well beyond the threshold of hearing, have nonlinearities artifacts that (periodically) increase in magnitude with respect to increase at ultrasonic frequencies. These nonlinearities essentially mask the radiation pressure generated because the magnitude of these nonlinearities are orders of magnitude greater than the radiation pressure. The acoustic (lowpass) filter referred to in this paper was designed in order to filter out the "detrimental" ultrasonic wave that was inducing high nonlinear noise artifacts in the measurement microphones. The highintensity, ultrasonic wave was producing radiation pressure (which is audible) within the initial acoustic medium (i.e. air). By filtering out the ultrasonic wave, the measurement microphone would only detect the audible radiation pressure that the ultrasonic wave was producing in air. Acoustic filters like these could possibly be used to detect/receive any highintensity, ultrasonic signal that may carry audio information which may need to be extrapolated with an acceptable level of fidelity.
References
[1] Minoru Todo, "New Type of Acoustic Filter Using Periodic Polymer Layers for Measuring Audio Signal Components Excited by AmplitudeModulated HighIntensity Ultrasonic Waves," Journal of Audio Engineering Society, Vol. 53, pp. 93041 (2005 October)
[2] Fundamentals of Acoustics; Kinsler et al., John Wiley & Sons, 2000
[3] ME 513 Course Notes, Dr. Luc Mongeau, Purdue University
[4] http://www.ieeeuffc.org/archive/uffc/trans/Toc/abs/02/t0270972.htm
Noise in Hydraulic Systems
Noise in Hydraulic Systems
Hydraulic systems are the most preferred source of power transmission in most of the industrial and mobile equipments due to their power density, compactness, flexibility, fast response and efficiency. The field hydraulics and pneumatics is also known as 'Fluid Power Technology'. Fluid power systems have a wide range of applications which include industrial, offroad vehicles, automotive system and aircrafts. But, one of the main problems with the hydraulic systems is the noise generated by them. The health and safety issues relating to noise have been recognized for many years and legislation is now placing clear demands on manufacturers to reduce noise levels [1]. Hence, noise reduction in hydraulic systems demands lot of attention from the industrial as well as academic researchers. It needs a good understanding of how the noise is generated and propagated in a hydraulic system in order to reduce it.
Sound in fluids
The speed of sound in fluids can be determined using the following relation.
where K  fluid bulk modulus,  fluid density, c  velocity of sound
Typical value of bulk modulus range from 2e9 to 2.5e9 N/m2. For a particular oil, with a density of 889 kg/m3,
speed of sound
Source of Noise
The main source of noise in hydraulic systems is the pump which supplies the flow. Most of the pumps used are positive displacement pumps. Of the positive dispalcement pumps, axial piston swash plate type is mostly preferred due to their controllability and efficiency.
The noise generation in an axial piston pump can be classifeid under two categories (i) fluidborne nose and
(ii) Structureborne noise
Fluidborne Noise (FBN)
Among the positive displacement pumps, highest levels of FBN are generated by axial piston pumps and lowest levels by screw pumps and in between these lie the external gear pump and vane pump [1]. The discussion in this page is mainly focused on axial piston swash plate type pumps. An axial piston pump has a fixed number of displacement chambers arranged in a circular pattern separated from each other by an angular pitch equal to where n is the number of displacement chambers. As each chamber discharges a specific volume of fluid, the discharge at the pump outlet is sum of all the discharge from the individual chambers. The discontinuity in flow between adjacent chambers results in a kinematic flow ripple. The amplitude of the kinematic ripple can be theoretically determined given the size of the pump and the number of displament chambers. The kinematic ripple is the main cause of the fluidborne noise. The kinematic ripples is a theoretical value. The actual flow ripple at the pump outlet is much larger than the theoretical value because the kinematic ripple is combined with a compressibility component which is due to the fluid compressibility. These ripples (also referred as flow pulsations) generated at the pump are transmitted through the pipe or flexible hose connected to the pump and travel to all parts of the hydraulic circuit.
The pump is considered an ideal flow source. The pressure in the system will be decided by resistance to the flow or otherwise known as system load. The flow pulsations result in pressure pulsations. The pressure pulsations are superimposed on the mean system pressure. Both the flow and pressure pulsations easily travel to all part of the circuit and affect the performance of the components like control valve and actuators in the system and make the component vibrate, sometimes even resonate. This vibration of system components adds to the noise generated by the flow pulsations. The transmission of FBN in the circuit is discussed under transmission below.
A typical axial piston pump with 9 pistons running at 1000 rpm can produce a sound pressure level of more than 70 dBs.
Structureborne Noise (SBN)
In swash plate type pumps, the main sources of the structureborne noise are the fluctuating forces and moments of the swas plate. These fluctuating forces arise as a result of the varying pressure inside the displacement chamber. As the displacing elements move from suction stroke to discharge stroke, the pressure varies accordingly from few bars to few hundred bars. This pressure changes are reflected on the displacement elements (in this case, pistons) as forces and these force are exerted on the swash plate causing the swash plate to vibrate. This vibration of the swash plate is the main cause of structureborne noise. There are other components in the system which also vibrate and lead to structureborne noise, but the swash is the major contributor.
Transmission
FBN
The transmission of FBN is a complex phenomenon. Over the past few decades, considerable amount of research had gone into mathematical modeling of pressure and flow transient in the circuit. This involves the solution of wave equations, with piping treated as a distributed parameter system known as a transmission line [1] & [3].
Lets consider a simple pumppipeloading valve circuit as shown in Fig. 2. The pressure and flow ripple at ay location in the pipe can be described by the relations:
where and are frequency dependent complex coefficients which are directly proportional to pump (source) flow ripple, but also functions of the source impedance , characteristic impedance of the pipe and the termination impedance . These impedances ,usually vary as the system operating pressure and flow rate changes, can be determined experimentally.
For complex systems with several system compenents, the pressure and flow ripples are estimated using the transformation matrix approach. For this, the system compenents can be treated as lumped impedances (a throttle valve or accumulator), or distributed impedances (flexible hose or silencer). Various software packages are available today to predict the pressure pulsations.
SBN
The transmission of SBN follows the classic sourcepathnoise model. The vibrations of the swash plate, the main cause of SBN, is transferred to the pump casing which encloses all the rotating group in the pump including displacement chambers (also known as cylinder block), pistons and the swash plate. The pump case, apart from vibrating itself, transfers the vibration down to the mount on which the pump is mounted. The mount then passes the vibrations down to the main mounted structure or the vehicle. Thus the SBN is transferred from the swash plate to the main strucuture or vehicle via pumpcasing and mount.
Some of the machine structures, along the path of transmission, are good at transmitting this vribational energy and they even resonate and reinforce it. By converting only a fraction of 1% of the pump structureborne noise into sound, a member in the transmission path could radiate more ABN than the pump itself [4].
Airborne noise (ABN)
Both FBN and SBN , impart high fatigue loads on the system components and make them vibrate. All of these vibrations are radiated as airborne noise and can be heard by a human operator. Also, the flow and pressure pulsations make the system components such as a control valve to resonate. This vibration of the particular component again radiates airborne noise.
Noise reduction
The reduction of the noise radiated from the hydraulic system can be approached in two ways.
(i) Reduction at Source  which is the reduction of noise at the pump. A large amount of open literature are availabale on the reduction techniques with some techniques focusing on reducing FBN at source and others focusing on SBN. Reduction in FBN and SBN at the source has a large influence on the ABN that is radiated. Even though, a lot of progress had been made in reducing the FBN and SBN separately, the problem of noise in hydraulic systems is not fully solved and lot need to be done. The reason is that the FBN and SBN are interlated, in a sense that, if one tried to reduce the FBN at the pump, it tends to affect the SBN characteristics. Currently, one of the main researches in noise reduction in pumps, is a systematic approach in understanding the coupling between FBN and SBN and targeting them simultaneously instead of treating them as two separte sources. Such an unified approach, demands not only well trained researchers but also sophisticated computer based mathematical model of the pump which can accurately output the necessary results for optimization of pump design. The amplitude of fluid pulsations can be reduced, at the source, with the use of an hydraulic attenuator(5).
(ii) Reduction at Component level  which focuses on the reduction of noise from individual component like hose, control valve, pump mounts and fixtures. This can be accomplished by a suitable design modification of the component so that it radiates least amount of noise. Optimization using computer based models can be one of the ways.
Hydraulic System noise
References
1. Designing Quieter Hydraulic Systems  Some Recent Developements and Contributions, Kevin Edge, 1999, Fluid Power: Forth JHPS International Symposium.
2. Fundamentals of Acoustics L.E. Kinsler, A.R. Frey, A.B.Coppens, J.V. Sanders. Fourth Edition. John Wiley & Sons Inc.
3. Reduction of Axial Piston Pump Pressure Ripple A.M. Harrison. PhD thesis, University of Bath. 1997
4. Noise Control of Hydraulic Machinery Stan Skaistis, 1988. MARCEL DEKKER , INC.
5. Hydraulic Power System Analysis, A. Akers, M. Gassman, & R. Smith, Taylor & Francis, New York, 2006, ISBN: 0824799569
Basic Acoustics of the Marimba
Introduction
One of my favorite instruments is the marimba. Like a xylophone, a marimba has octaves of wooden bars that are struck with mallets to produce tones. Unlike the harsh sound of a xylophone, a marimba produces a deep, rich tone. Marimbas are not uncommon and are played in most high school bands.
Now, while all the trumpet and flute and clarinet players are busy tuning up their instruments, the marimba player is back in the percussion section with her feet up just relaxing. This is a bit surprising, however, since the marimba is a melodic instrument that needs to be in tune to sound good. So what gives? Why is the marimba never tuned? How would you even go about tuning a marimba? To answer these questions, the acoustics behind (or within) a marimba must be understood.
Components of Sound
What gives the marimba its unique sound? It can be boiled down to two components: the bars and the resonators. Typically, the bars are made of rosewood (or some synthetic version of wood). They are cut to size depending on what note is desired, then the tuning is refined by shaving wood from the underside of the bar.
Example: Rosewood bar, middle C, 1 cm thick
The equation that relates the length of the bar with the desired frequency comes from the theory of modeling a bar that is free at both ends. This theory yields the following equation:
where t is the thickness of the bar, v is the speed of sound in the bar, and f is the frequency of the note. For rosewood, v = 5217 m/s. For middle C, f=262 Hz. Therefore, to make a middle C key for a rosewood marimba, cut the bar to be:

The resonators are made from metal (usually aluminum) and their lengths also differ depending on the desired note. It is important to know that each resonator is open at the top but closed by a stopper at the bottom end.
Example: Aluminum resonator, middle C
The equation that relates the length of the resonator with the desired frequency comes from modeling the resonator as a pipe that is driven at one end and closed at the other end. A "driven" pipe is one that has a source of excitation (in this case, the vibrating key) at one end. This model yields the following:
where c is the speed of sound in air and f is the frequency of the note. For air, c = 343 m/s. For middle C, f = 262 Hz. Therefore, to make a resonator for the middle C key, the resonator length should be:

Resonator Shape
The shape of the resonator is an important factor in determining the quality of sound that can be produced. The ideal shape is a sphere. This is modeled by the Helmholtz resonator. (For more see Helmholtz Resonator page) However, mounting big, round, beach balllike resonators under the keys is typically impractical. The worst choices for resonators are square or oval tubes. These shapes amplify the nonharmonic pitches sometimes referred to as “junk pitches”. The round tube is typically chosen because it does the best job (aside from the sphere) at amplifying the desired harmonic and not much else.
As mentioned in the second example above, the resonator on a marimba can be modeled by a closed pipe. This model can be used to predict what type of sound (full and rich vs dull) the marimba will produce. Each pipe is a "quarter wave resonator" that amplifies the sound waves produced by of the bar. This means that in order to produce a full, rich sound, the length of the resonator must exactly match onequarter of the wavelength. If the length is off, the marimba will produce a dull or offkey sound for that note.
Why would the marimba need tuning?
In the theoretical world where it is always 72 degrees with low humidity, a marimba would not need tuning. But, since weather can be a factor (especially for the marching band) marimbas do not always perform the same way. Hot and cold weather can wreak havoc on all kinds of percussion instruments, and the marimba is no exception. On hot days, the marimba tends to be sharp and for cold days it tends to be flat. This is the exact opposite of what happens to string instruments. Why? The tone of a string instrument depends mainly on the tension in the string, which decreases as the string expands with heat. The decrease in tension leads to a flat note. Marimbas on the other hand produce sound by moving air through the resonators. The speed at which this air is moved is the speed of sound, which varies proportionately with temperature! So, as the temperature increases, so does the speed of sound. From the equation given in example 2 from above, you can see that an increase in the speed of sound (c) means a longer pipe is needed to resonate the same note. If the length of the resonator is not increased, the note will sound sharp. Now, the heat can also cause the wooden bars to expand, but the effect of this expansion is insignificant compared to the effect of the change in the speed of sound.
Tuning Myths
It is a common myth among percussionists that the marimba can be tuned by simply moving the resonators up or down (while the bars remain in the same position.) The thought behind this is that by moving the resonators down, for example, you are in effect lengthening them. While this may sound like sound reasoning, it actually does not hold true in practice. Judging by how the marimba is constructed (cutting bars and resonators to specific lengths), it seems that there are really two options to consider when looking to tune a marimba: shave some wood off the underside of the bars, or change the length of the resonator. For obvious reasons, shaving wood off the keys every time the weather changes is not a practical solution. Therefore, the only option left is to change the length of the resonator. As mentioned above, each resonator is plugged by a stopper at the bottom end. So, by simply shoving the stopper farther up the pipe, you can shorten the resonator and sharpen the note. Conversely, pushing the stopper down the pipe can flatten the note. Most marimbas do not come with tunable resonators, so this process can be a little challenging. (Broomsticks and hammers are common tools of the trade.)
Example: Middle C Resonator lengthened by 1 cm
For ideal conditions, the length of the middle C (262 Hz) resonator should be 32.7 cm as shown in example 2. Therefore, the change in frequency for this resonator due to a change in length is given by:
If the length is increased by 1 cm, the change in frequency will be:

The acoustics behind the tuning a marimba go back to the design that each resonator is to be ¼ of the total wavelength of the desired note. When marimbas get out of tune, this length is no longer exactly equal to ¼ the wavelength due to the lengthening or shortening of the resonator as described above. Because the length has changed, resonance is no longer achieved, and the tone can become muffled or offkey.
Conclusions
Some marimba builders are now changing their designs to include tunable resonators. Since any leak in the endseal will cause major loss of volume and richness of the tone, this is proving to be a very difficult task. At least now, though, armed with the acoustic background of their instruments, percussionists everywhere will now have something to do when the conductor says, “tune up!”
Links and References
 http://www.gppercussion.com/html/resonators.html
 http://www.mostlymarimba.com/
 http://www.craftymusicteachers.com/bassmarimba/
How an Acoustic Guitar works
Introduction
There are three main parts of the guitar that contribute to sound production. First of all, there are the strings. Any string that is under tension will vibrate at a certain frequency. The tension and gauge in the string determine the frequency at which it vibrates. The guitar controls the length and tension of six differently weighted strings to cover a very wide range of frequencies. Second, there is the body of the guitar. The guitar body is connected directly to one end of each of the strings. The body receives the vibrations of the strings and transmits them to the air around the body. It is the body’s large surface area that allows it to “push” a lot more air than a string. Finally, there is the air inside the body. This is very important for the lower frequencies of the guitar. The air mass just inside the sound hole oscillates, compressing and decompressing the compliant air inside the body. In practice this concept is called a Helmholtz resonator. Without this, it would difficult to produce the wonderful timbre of the guitar.
The Strings
The strings of the guitar vary in linear density, length, and tension. This gives the guitar a wide range of attainable frequencies. The larger the linear density is, the slower the string vibrates. The same goes for the length; the longer the string is the slower it vibrates. This causes a low frequency. Inversely, if the strings are less dense and/or shorter they create a higher frequency. The lowest resonance frequencies of each string can be calculated by
where = string tension, =linear density, = string length
The string length, L, in the equation is what changes when a player presses on a string at a certain fret. This will shorten the string which in turn increases the frequency it produces when plucked. The spacing of these frets is important. The length from the nut to bridge determines how much space goes between each fret. If the length is 25 inches, then the position of the first fret should be located (25/17.817) inches from the nut. Then the second fret should be located (25(25/17.817))/17.817 inches from the first fret. This results in the equation
When a string is plucked, a disturbance is formed and travels in both directions away from point where the string was plucked. These "waves" travel at a speed that is related to the tension and linear density and can be calculated by
The waves travel until they reach the boundaries on each end where they are reflected back. The link below displays how the waves propagate in a string.
Plucked String @ www.phys.unsw.edu
The strings themselves do not produce very much sound because they are so thin. This is why they are connected to the top plate of the guitar body. They need to transfer the frequencies they are producing to a large surface area which can create more intense pressure disturbances.
The Body
The body of the guitar transfers the vibrations of the bridge to the air that surrounds it. The top plate contributes to most of the pressure disturbances, because the player dampens the back plate and the sides are relatively stiff. This is why it is important to make the top plate out of a light springy wood, like spruce. The more the top plate can vibrate, the louder the sound it produces will be. It is also important to keep the top plate flat, so a series of braces are located on the inside to strengthen it. Without these braces the top plate would bend and crack under the large stress created by the tension in the strings. This would also affect the magnitude of the sound being transmitted. The warped plate would not be able to "push" air very efficiently. A good experiment to try, in order to see how important this part of the guitar is in the amplification process, is as follows:
1. Start with an ordinary rubber band, a large bowl, adhesive tape, and plastic wrap.
2. Stretch the rubber band and pluck it a few times to get a good sense for how loud it is.
3. Stretch the plastic wrap over the bowl to form a sort of drum.
4. Tape down one end of the rubber band to the plastic wrap.
5. Stretch the rubber band and pluck it a few times.
6. The sound should be much louder than before.
The Air
The final part of the guitar is the air inside the body. This is very important for the lower range of the instrument. The air just inside the soundhole oscillates compressing and expanding the air inside the body. This is just like blowing across the top of a bottle and listening to the tone it produces. This forms what is called a Helmholtz resonator. For more information on Helmholtz resonators go to Helmholtz Resonance. This link also shows the correlation to acoustic guitars in great detail. The acoustic guitar makers often tune these resonators to have a resonance frequency between F#2 and A2 (92.5 to 110.0 Hz). Having such a low resonance frequency is what aids the amplification of the lower frequency strings. To demonstrate the importance of the air in the cavity, simply play an open A on the guitar (the second string). Now, as the string is vibrating, place a piece of cardboard over the soundhole. The sound level is reduced dramatically. This is because you've stopped the vibration of the air mass just inside the soundhole, causing only the top plate to vibrate. Although the top plate still vibrates a transmits sound, it isn't as effective at transmitting lower frequency waves, thus the need for the Helmholtz resonator.
Specific applicationautomobile muffler
General information about Automobile muffler
Introduction
A muffler is a part of the exhaust system on an automobile that plays a vital role. It needs to have modes that are located away from the frequencies that the engine operates at, whether the engine be idling or running at the maximum amount of revolutions per second.A muffler that affects an automobile in a negative way is one that causes noise or discomfort while the car engine is running.Inside a muffler, you'll find a deceptively simple set of tubes with some holes in them. These tubes and chambers are actually as finely tuned as a musical instrument. They are designed to reflect the sound waves produced by the engine in such a way that they partially cancel themselves out.( cited from www.howstuffworks.com )
It is very important to have it on the automobile. The legal limit for exhaust noise in the state of California is 95dB (A)  CA. V.C. 27151 .Without a muffler the typical car exhaust noise would exceed 110dB.A conventional car muffler is capable of limiting noise to about 90 dB. The activenoise canceling muffler enables cancellation of exhaust noise to a wide range of frequencies.
The Configuration of A automobile muffler
How Does automobile muffler function?
General Concept
The simple and main part of designing the automobile muffler is to use the lowpass filter. It typically makes use of the change of the cross section area which can be made as a chamber to filter or reduce the sound wave which the engine produced.
LowPass Filter
A lowpass filter is a circuit that passes low frequency signals but stops the high frequency signals. Once the low pass filter is set by the user at a specific cutoff frequency, all frequencies lower than that will be passed through the filter, while higher frequencies will be attenuated in amplitude. This circuit made up of passive components (resistor, capacitors and inductors) capable of accomplishing this objective. File:Inductive law pass filter.jpg
the formula to be used:
Human ear sound reaction feature
When these pressure pulses reach your ear, the eardrum vibrates back and forth. Your brain interprets this motion as sound. Two main characteristics of the wave determine how we perceive the sound:
1.sound wave frequency. 2.air wave pressure amplitude.
It turns out that it is possible to add two or more sound waves together and get less sound.
Description of the muffler to cancel the noise
The key thing about sound waves is that the result at your ear is the sum of all the sound waves hitting your ear at that time. If you are listening to a band, even though you may hear several distinct sources of sound, the pressure waves hitting your ear drum all add together, so your ear drum only feels one pressure at any given moment. Now comes the cool part: It is possible to produce a sound wave that is exactly the opposite of another wave. This is the basis for those noisecanceling headphones you may have seen. Take a look at the figure below. The wave on top and the second wave are both pure tones. If the two waves are in phase, they add up to a wave with the same frequency but twice the amplitude. This is called constructive interference. But, if they are exactly out of phase, they add up to zero. This is called destructive interference. At the time when the first wave is at its maximum pressure, the second wave is at its minimum. If both of these waves hit your ear drum at the same time, you would not hear anything because the two waves always add up to zero.
Benefits of an Active NoiseCanceling Muffler
1.By using an active muffler the exhaust noise can be easily tuned, amplified, or nearly eliminated.
2.The backpressure of a conventional muffler can be essentially eliminated, thus increasing engine performance and efficiency.
3.By increasing engine efficiency and performance, less fuel will be used and the emissions will be reduced.
Absorptive muffler
Lined ducts
It can be regarded as simplest form of absorptive muffler. Attach absorptive material to the bare walls of the duct.( in car that is the exhaustion tube) The attenuation performance improves with the thickness of absorptive material.
The attenuation curves like a skewed bell. Increase the thickness of the wall will get the lower maximum attenuation frequency. For higher frequency though, thinner absorbent layers are effective, but the large gap allows noise to pass directly along. Thin layers and narrow passages are therefore more effective at high frequencies. For good absorption over the widest frequency range, thick absorbent layers and narrow passages are best.
Parallel and blocklineofsight baffles
Divide the duct into several channels or turn the flow channels so that there is no direct lineofsight through the baffles. Frequently the materials line on the channels. Attenuation improves with the thickness of absorptive material and length of the baffle. Lined bends can be used to provide a greater attenuation and attenuate best at high frequency. Comparatively, at low frequency attenuation can be increased by adding thicker lining.
Plenum chambers
They are relatively large volume chambers, usually fabricated from sheet metal, which interconnect two ducts. The interior of the chamber is lined with absorbing material to attenuate noise in the duct. Protective facing material may also be necessary if the temperature and velocity conditions of the gas stream are too severe.
The performance of a plenum chamber can be improved by: 1.increase the thickness of the absorbing lining 2.blocking the direct line of sight from the chamber inlet to the outlet. 3.increase the crosssectional area of the chamber.
Bessel Functions and the Kettledrum
Abstract
In class, we have begun to discuss the solutions of multidimentional wave equations. A particularly interesting aspect of these multidimentional solutions are those of bessel functions for circular boundary condictions. The practical application of these solutions is the kettledrum. This page will explore in qualitative and quantitative terms how the of the kettledrum works. More specifically, the kettledrum will be introduced as a circular membrane and it solution will be discussed in visual (e.g. visualization of bessel functions, video of kettledrums and audio forms (wav files of kettledrums playing. In addition, links to more information about this material, including references will be included.
What is a kettledrum
A kettledrum is a percussion instrument with a circular drumhead mounted on a "kettlelike" enclosure. When one strikes the drumhead with a mallet, it vibrates which produces its sound. The pitch of this sound is determined by the tension of the drumhead, which is precisely tuned before playing. The sound of the kettldrum (called the Timpani in classical music) is present in many forms of music from many difference places of the world.
The math behind the kettledrum:the brief version
When one looks at how a kettledrum produces sound, one should look no farther than the drumhead. The vibration of this circular membrane (and the air in the drum enclosure) is what produces the sound in this instrument. The mathematics behind this vibrating drum are relatively simple. If one looks at a small element of the drum head, it looks exactly like the situation for the vibrating string (see:). The only difference is that there are two dimensions where there are forces on the element, the two dimensions that are planar to the drum. As this is the same situation, we have the same equation, except with another spatial term in the other planar dimension. This allows us to model the drumhead using a helmholtz equation. The next step (solved in detail below) is to assume that the displacement of the drumhead (in polar coordinates) is a product of two separate functions for theta and r. This allows us to turn the PDE into two ODES which are readily solved and applied to the situation of the kettledrum head. For more info, see below.
The math behind the kettledrum:the derivation
So starting with the trusty general Helmholtz equation:
Where k is the wave number, the frequency of the forced oscillations divided by the speed of sound in the membrane.
Since we are dealing with a circular object, it make sense to work in polar coordinates (in terms of radius and angle) instead of rectangular coordinates. For polar coordinates the Laplacian term of the helmholtz relation () becomes
Now lets assume that:
This assumption follows the method of separation of variables. (see Reference 3 for more info) Substituting this result back into our trusty Helmholtz equation gives the following:
Since we separated the variables of the solution into two onedimensional functions, the partial dirivatives become ordinary dirivatives. Both sides of this result must equal the same constant. For simplicity, i will use as this constant. This results in the following two equations:
The first of these equations readily seen as the standard second order ordinary differencial equation which has a harmonic solution of sines and cosines with the frequency based on . The second equation is what is known as Bessel's Equation. The solution to this equation is cryptically called Bessel functions of order of the first and second kind. These functions, while sounding very intimidating, are simply oscillatory functions of the radius times the wave number that are unbounded at when kr (for the function of the second kind) approaches zero and diminish as kr get larger. (For more information on what these functions look like see References 1,2, and 3)
Now that we have the general solution to this equation, we can now model a infinite radius kettledrum head. However, since i have yet to see an infinite kettle drum, we need to constrain this solution of a vibrating membrane to a finite radius. We can do this by applying what we know about our circular membrane: along the edges of the kettledrum, the drum head is attached to the drum. This means that there can be no displacement of the membrane at the termination at the radius of the kettle drum. This boundary condiction can be mathematically described as the following:
Where a is the arbirary radius of the kettledrum. In addition to this boundary condiction, the displacement of the drum head at the center must be finite. This second boundary condiction removes the bessel function of the second kind from the solution. This reductes the R part of our solution to:
Where is a bessel function of the first kind of order . Apply our other boundary condiction at the radius of the drum requires that the wave number k must have discrete values, () which can be looked up. Combining all of these gives us our solution to how a drumhead behaves (which is the real part of the following):
The math behind the kettledrum:the entire drum
The above derivation is just for the drum head. An actual kettledrum has one side of this circular membrane surrounded by an enclosed cavity. This means that air is compressed in the cavity when the membrane is vibrating, adding more complications to the solution. In mathematical terms, this makes the partial differencial equation nonhomogeneous or in simpler terms, the right side of the Helmholtz equation does not equal zero. This result requires significantly more derivation, and will not be done here. If the reader cares to know more, these results are discussed in the two books under references 6 and 7.
Sites of interest
As one can see from the derivation above, the kettledrum is very interesting mathematically. However, it also has a rich historical music tradition in various places of the world. As this page's emphasis is on math, there are few links provided below that reference this rich history.
A discussion of Persian kettledrums: Kettle drums of Iran and other countries
A discussion of kettledrums in classical music: Kettle drum Lit.
A massive resource for kettledrum history, construction and technique" Vienna Symphonic Library
Wikibooks sister cite, references under Timpani: Wikipedia reference
References
1.Eric W. Weisstein. "Bessel Function of the First Kind." From MathWorldA Wolfram Web Resource. http://mathworld.wolfram.com/BesselFunctionoftheFirstKind.html
2.Eric W. Weisstein. "Bessel Function of the Second Kind." From MathWorldA Wolfram Web Resource. http://mathworld.wolfram.com/BesselFunctionoftheSecondKind.html
3.Eric W. Weisstein. "Bessel Function." From MathWorldA Wolfram Web Resource. http://mathworld.wolfram.com/BesselFunction.html
4.Eric W. Weisstein et al. "Separation of Variables." From MathWorldA Wolfram Web Resource. http://mathworld.wolfram.com/SeparationofVariables.html
5.Eric W. Weisstein. "Bessel Differential Equation." From MathWorldA Wolfram Web Resource. http://mathworld.wolfram.com/BesselDifferentialEquation.html
6. Kinsler and Frey, "Fundamentals of Acoustics", fourth edition, Wiley & Sons
7. Haberman, "Applied Partial Differential Equations", fourth edition, Prentice Hall Press
Filter Design and Implementation
Introduction
Acoustic filters, or mufflers, are used in a number of applications requiring the suppression or attenuation of sound. Although the idea might not be familiar to many people, acoustic mufflers make everyday life much more pleasant. Many common appliances, such as refrigerators and air conditioners, use acoustic mufflers to produce a minimal working noise. The application of acoustic mufflers is mostly directed to machine components or areas where there is a large amount of radiated sound such as high pressure exhaust pipes, gas turbines, and rotary pumps.
Although there are a number of applications for acoustic mufflers, there are really only two main types which are used. These are absorptive and reactive mufflers. Absorptive mufflers incorporate sound absorbing materials to attenuate the radiated energy in gas flow. Reactive mufflers use a series of complex passages to maximize sound attenuation while meeting set specifications, such as pressure drop, volume flow, etc. Many of the more complex mufflers today incorporate both methods to optimize sound attenuation and provide realistic specifications.
In order to fully understand how acoustic filters attenuate radiated sound, it is first necessary to briefly cover some basic background topics. For more information on wave theory and other material necessary to study acoustic filters please refer to the references below.
Basic Wave Theory
Although not fundamentally difficult to understand, there are a number of alternate techniques used to analyze wave motion which could seem overwhelming to a novice at first. Therefore, only 1D wave motion will be analyzed to keep most of the mathematics as simple as possible. This analysis is valid, with not much error, for the majority of pipes and enclosures encountered in practice.
PlaneWave Pressure Distribution in Pipes
The most important equation used is the wave equation in 1D form (See [1],[2], [1D Wave Equation], [Vibrations of Strings] for information).
Therefore, it is reasonable to suggest, if plane waves are propagating, that the pressure distribution in a pipe is given by:
where Pi and Pr are incident and reflected wave amplitudes respectively. Also note that bold notation is used to indicate the possiblily of complex terms. The first term represents a wave travelling in the +x direction and the second term, x direction.
Since acoustic filters or mufflers typically attenuate the radiated sound power as much as possible, it is logical to assume that if we can find a way to maximize the ratio between reflected and incident wave amplitude then we will effectively attenuated the radiated noise at certain frequencies. This ratio is called the reflection coefficient and is given by:
It is important to point out that wave reflection only occurs when the impedance of a pipe changes. It is possible to match the end impedance of a pipe with the characteristic impedance of a pipe to get no wave reflection. For more information see [1] or [2].
Although the reflection coefficient isn't very useful in its current form since we want a relation describing sound power, a more useful form can be derived by recognizing that the power intensity coefficient is simply the magnitude of reflection coefficient square [1]:
As one would expect, the power reflection coefficient must be less than or equal to one. Therefore, it is useful to define the transmission coefficient as it said:
which is the amount of power transmitted. This relation comes directly from conservation of energy. When talking about the performance of mufflers, typically the power transmission coefficient is specified.
Basic Filter Design
For simple filters, a long wavelength approximation can be made to make the analysis of the system easier. When this assumption is valid (e.g. low frequencies) the components of the system behave as lumped acoustical elements. Equations relating the various properties are easily derived under these circumstances, see [Lumped Elements] for further information.
The following derivations assume long wavelength. Practical applications for most conditions are given later.
LowPass Filter
These are devices that attenuate the radiated sound power at higher frequencies. This means the power transmission coefficient is approximently 1 across the band pass at low frequencies(see figure to right).
This is equivalent to an expansion in a pipe, with the volume of gas located in the expansion having an acoustic compliance (see figure to right). Continuity of acoustic impedance (see Java Applet at: [Acoustic Impedance Visualization]) at the junction, see [1], gives a power transmission coefficient of:
where k is the wavenumber (see [Wave Properties]), L & are length and area of expansion respectively, and S is the area of the pipe.
The cutoff frequency is given by:
HighPass Filter
These are devices that attenuate the radiated sound power at lower frequencies. Like before, this means the power transmission coefficient is approximately 1 across the band pass at high frequencies (see figure to right).
This is equivalent to a short side branch (see figure to right) with a radius and length much smaller than the wavelength (lumped element assumption). This side branch acts like an acoustic mass and applies a different acoustic impedance to the system than the lowpass filter. Again using continuity of acoustic impedance at the junction yields a power transmission coefficient of the form [1]:
where a and L are the radius and effective length of the small tube, respectively, and S is the area of the pipe.
The cutoff frequency is given by:
BandStop Filter
These are devices that attenuate the radiated sound power over a certain frequency range (see figure to right). Like before, the power transmission coefficient is approximently 1 in the band pass region.
Since the bandstop filter is essentially a cross between a low and high pass filter, one might expect to create one by using a combination of both techniques. This is true in that the combination of a lumped acoustic mass and compliance gives a bandstop filter. This can be realized as a helmholtz resonator (see [Helmholtz Resonator] or figure to right). Again, since the impedance of the helmholtz resonator can be easily determined, continuity of acoustic impedance at the junction can give the power transmission coefficient as [1]:
where is the area of the neck, L is the effective length of the neck, V is